Band 65, Heft 4

The general uncertainty of epilepsy and its unpredictable seizures often affect badly the quality of life of people exposed to this disease. There are patients who can be considered fortunate in terms of prediction of any seizures. These are patients with epileptic auras. In this study, it was aimed to evaluate pre-seizure warning symptoms of the electroencephalography (EEG) signals by a convolutional neural network (CNN) inspired by the epileptic auras defined in the medical field. In this context, one-dimensional EEG signals were transformed into a spectrogram display form in the frequency-time domain by applying a short-time Fourier transform (STFT). Systemic changes in pre-epileptic seizure have been described by applying the CNN approach to the EEG signals represented in the image form, and the subjective EEG-Aura process has been tried to be determined for each patient. Considering all patients included in the evaluation, it was determined that the 1-min interval covering the time from the second minute to the third minute before the seizure had the highest mean and the lowest variance to determine the systematic changes before the seizure. Thus, the highest performing process is described as EEG-Aura. The average success for the EEG-Aura process was 90.38 ± 6.28%, 89.78 ± 8.34% and 90.47 ± 5.95% for accuracy, specificity and sensitivity, respectively. Through the proposed model, epilepsy patients who do not respond to medical treatment methods are expected to maintain their lives in a more comfortable and integrated way.

In this paper, we suggest an efficient, accurate and user-friendly brain-computer interface (BCI) system for recognizing and distinguishing different emotion states. For this, we used a multimodal dataset entitled “MAHOB-HCI” which can be freely reached through an email request. This research is based on electroencephalogram (EEG) signals carrying emotions and excludes other physiological features, as it finds EEG signals more reliable to extract deep and true emotions compared to other physiological features. EEG signals comprise low information and signal-to-noise ratios (SNRs); so it is a huge challenge for proposing a robust and dependable emotion recognition algorithm. For this, we utilized a new method, based on the matching pursuit (MP) algorithm, to resolve this imperfection. We applied the MP algorithm for increasing the quality and SNRs of the original signals. In order to have a signal of high quality, we created a new dictionary including 5-scale Gabor atoms with 5000 atoms. For feature extraction, we used a 9-scale wavelet algorithm. A 32-electrode configuration was used for signal collection, but we used just eight electrodes out of that; therefore, our method is highly user-friendly and convenient for users. In order to evaluate the results, we compared our algorithm with other similar works. In average accuracy, the suggested algorithm is superior to the same algorithm without applying MP by 2.8% and in terms of f-score by 0.03. In comparison with corresponding works, the accuracy and f-score of the proposed algorithm are better by 10.15% and 0.1, respectively. So as it is seen, our method has improved past works in terms of accuracy, f-score and user-friendliness despite using just eight electrodes.

Wearable smart monitors (WSMs) applied for the estimation of electrophysiological signals are of utmost interest for a non-stressed life. WSM which records heart muscle activities could signalize timely a life-threatening event. The heart muscle activities are typically recorded across the heart at the surface of the body; hence, a WSM monitor requires high-quality surface electrodes. The electrodes used in the clinical settings [i.e. silver/silver chloride (Ag/AgCl) with the gel] are not practical for the daily out of clinic usage. A practical WSM requires the application of a dry electrode with stable and reproducible electrical characteristics. We compared the characteristics of six types of dry electrodes and one gelled electrode during short-term recordings sessions (≈30 s) in real-life conditions: Orbital, monolithic polymer plated with Ag/AgCl, and five rectangular shaped 10 × 6 × 2 mm electrodes (Orbital, Ag electrode, Ag/AgCl electrode, gold electrode and stainless-steel AISI304). The results of a well-controlled analysis which considered motion artifacts, line noise and junction potentials suggest that among the dry electrodes Ag/AgCl performs the best. The Ag/AgCl electrode is in average three times better compared with the stainless-steel electrode often used in WSMs.

A noninvasive investigation to ascertain the platelet (PLT) count was conducted on 44 hospitalized dengue hemorrhagic fever (DHF) subjects, male and female aged between 3 and 14 years using bioelectrical phase angle (BPhA). Among the 44 subjects, 30 subjects were confirmed to be non-structural protein-1 (NS1) positive at the time of admission, whose blood investigations such as hematocrit (HCT) level, PLT count, aspartate aminotransferase (AST) level and alanine aminotransferase (ALT) level were performed for the classification of risk as low-risk (LR) and high-risk (HR) DHF. It was found that the BPhA of the body reflects a linear correlation with the PLT count. To provide a better and more accurate estimate of PLT, a dual-frequency method is proposed to calculate the phase angle of the total body. The resistance at 5 kHz and reactance at 100 kHz were used to estimate the phase angle of the total body. The statistical analysis identified that the PLT count estimated using the proposed dual-frequency method shows a good correlation with the blood investigation results. In addition, statistical analysis of the proposed method on other fever subjects indicated a significant difference with DHF.

In this paper, a method for evaluating the chronological age of adolescents on the basis of their voice signal is presented. For every examined child, the vowels a, e, i, o and u were recorded in extended phonation. Sixty voice parameters were extracted from each recording. Voice recordings were supplemented with height measurement in order to check if it could improve the accuracy of the proposed solution. Predictor selection was performed using the LASSO (least absolute shrinkage and selection operator) algorithm. For age estimation, the random forest (RF) for regression method was employed and it was tested using a 10-fold cross-validation. The lowest absolute error (0.37 year ± 0.28) was obtained for boys only when all selected features were included into prediction. In all cases, the achieved accuracy was higher for boys than for girls, which results from the fact that the change of voice with age is larger for men than for women. The achieved results suggest that the presented approach can be employed for accurate age estimation during rapid development in children.

Intensive care units (ICUs) are responsible for generating a wealth of useful data in the form of electronic health records. We aimed to build a mortality prediction model on a Medical Information Mart for Intensive Care (MIMIC-III) database and to assess whether the use of deep learning techniques like long short-term memory (LSTM) can effectively utilize the temporal relations among clinical variables. The models were built on clinical variable dynamics of the first 48 h of ICU admission of 12,550 records from the MIMIC-III database. A total of 36 variables including 33 time series variables and three static variables were used for the prediction. We present the application of LSTM and LSTM attention (LSTM-AT) model for mortality prediction with such a large number of clinical variables dataset. For training and validation purpose, we have used International Classification of Diseases, 9th edition (ICD-9) codes for extracting the patients with cardiovascular disease, and infections and parasitic disease, respectively. The effectiveness of the LSTM model is achieved over non-recurrent baseline models like naïve Bayes, logistic regression (LR), support vector machine and multilayer perceptron (MLP) by generating state of the art results (area under the curve [AUC], 0.852). Next, by providing attention at each time stamp, we developed a model, LSTM-AT, which exhibits even better performance (AUC, 0.876).

Coordination and timing in repetitive movements have been intensively investigated in diverse experimental settings for understanding the underlying basic mechanisms in healthy controls. On this basic research side, there are mainly two theoretical models: the Wing-Kristofferson (WK) model and the Haken-Kelso-Bunz (HKB) model. On the clinical side of the research, several efforts have been spent on quantitatively assessing gait and other repetitive movements such as tapping, especially as an outcome measure of clinical trials in diverse neurological disorders. Nevertheless, Parkinson’s disease (PD) remains the predominant disorder in the clinical literature in this context, as the tremor activity and the changes in the gait are both common symptoms in PD. Although there are motion recording systems for data acquisition in clinical settings, the tools for analysis and quantification of the extracted time-series offered by these systems are severely restricted. Therefore, we introduce a toolbox which enables the analysis of repetitive movements within the framework of the two main theoretical models of motor coordination, which explicitly focuses on varying clinical and experimental settings such as self-paced vs. cued or uni-manual vs. bi-manual measurements. The toolbox contains particular pipelines for digital signal processing. Licensed under the GNU General Public License (GNU-GPL), the open source toolbox is freely available and can be downloaded from the Github link: https://github.com/MehmetEylemKirlangic/RepetitiveMovementAnalysis. We illustrate the application of the toolbox on sample experiments of gait and tapping with a control subject, as well as with a Parkinson’s patient. The patient has gone through a brain surgery for deep brain stimulation (DBS); hence, we present the results for both stimulation ON and stimulation OFF modes. Sample data are freely accessible at: https://github.com/MehmetEylemKirlangic/DATA.

Functional electrical stimulation (FES) has been used to produce force-related activities on the paralyzed muscle among spinal cord injury (SCI) individuals. Early muscle fatigue is an issue in all FES applications. If not properly monitored, overstimulation can occur, which can lead to muscle damage. A real-time mechanomyography (MMG)-based FES system was implemented on the quadriceps muscles of three individuals with SCI to generate an isometric force on both legs. Three threshold drop levels of MMG-root mean square (MMG-RMS) feature (thr50, thr60, and thr70; representing 50%, 60%, and 70% drop from initial MMG-RMS values, respectively) were used to terminate the stimulation session. The mean stimulation time increased when the MMG-RMS drop threshold increased (thr50: 22.7 s, thr60: 25.7 s, and thr70: 27.3 s), indicating longer sessions when lower performance drop was allowed. Moreover, at thr70, the torque dropped below 50% from the initial value in 14 trials, more than at thr50 and thr60. This is a clear indication of muscle fatigue detection using the MMG-RMS value. The stimulation time at thr70 was significantly longer (p = 0.013) than that at thr50. The results demonstrated that a real-time MMG-based FES monitoring system has the potential to prevent the onset of critical muscle fatigue in individuals with SCI in prolonged FES sessions.

The present study aimed to evaluate the characteristic influence of physical therapeutic procedures of vacuum-compression therapy (VCT) on microvascular perfusion (MiP) and macrovascular perfusion (MaP) of the lower limb in diabetic patients. A sample of nine patients with a medical history of type 2 diabetes was used for the purpose of this study. Most of the subjects’ medical conditions included venous and neurological complications of the lower limb, whereas the rest of the subjects entered the treatment due to injury recovery or their phlebological disease. The PeriFlux System 5000 (Perimed, Sweden) diagnostic device was used to measure MiP. The MaP was evaluated based on the perfusion index (PI) using the Extremiter monitoring device (Embitron, Czech Republic) designed to perform VCT procedures. The study found that MiP and MaP increase as an effect of VCT procedures and at the same time PI clearly reflects the effect of the applied vacuum and compression phases, verifying the method’s vital influence on peripheral perfusion disorders.

Threaded cups show good clinical results when implanted correctly. In clinical use, multiple cases with an incomplete placement of the EcoFit threaded cup (implantcast) were observed. This behaviour could not be explained intra- and postoperatively. The aim of this study was to compare and optimise the drill-in-behaviour of the EcoFit cup in a biomechanical investigation. EcoFit cup sizes 46, 50 and 54 mm were compared with the SC cup (Aesculap) size 50 mm. Foam blocks (Sawbones) of density 0.16 g/ml (pcf 10), 0.32 g/ml (pcf 20) and 0.48 g/ml (pcf 30) were used. After standardised placement using a universal testing system (n = 8 per group), the primary stability, the overhang of the cups and the drill-in behaviour were measured. Overreamings of 1 and 2 mm were performed (pcf 20, n = 8) for the EcoFit cup size 50 and the primary stability as well as the overhang measurements were examined. Measurements of the cup diameter, thread depth and thread pitch were performed on three-dimensional (3D) images of the cup size 50 mm. The drill-in behaviour was different between the EcoFit and the SC cups. Even with maximum torque, the EcoFit cup could not be positioned as deep as the SC cup in standard reaming conditions (overhang of 1.1 ± 0.4 mm for the EcoFit size 50 in pcf 20 and of −0.01 ± 0.2 mm for the SC cup). The primary stability was lower for the EcoFit cup in comparison to the SC cup (128.8 ± 3.2 Nm vs. 138.6 ± 9.1 Nm, p = 0.0291). With overreaming to 51 mm, a deeper positioning of the EcoFit was possible (overlap of −0.3 ± 0.1, comparable to the SC cup). The overreaming of the cavity also led to a significantly higher primary stability of 143.4 ± 3.7 Nm (p < 0.001) comparable to the unaltered condition (128.8 ± 3.2 Nm). Overreaming to 52 mm had no further advantage in terms of primary stability or overhang. The geometric measurements showed significant differences as well. The previously clinically observed difficulties in inserting the cup were confirmed by this study. By overreaming to 51 mm, the drill-in behaviour, the primary stability and the measured overhang were comparable to the reference cup. The obtained results suggest that the extension of the acetabulum cavity to 51 mm while using the implantcast EcoFit size 50 should be implemented in clinical applications.

This study presents a device for efficient, low-cost and eye-friendly overnight disinfection of contact lenses by visible violet light as an alternative to disinfection with biocide-containing solutions. Bacterial solutions with one Pseudomonas and one Staphylococcus strain each were irradiated for up to 8 h in commercial transparent contact lens cases by the presented light-emitting diode (LED) device. Samples were taken at different intervals and distributed on agar plates. The surviving bacteria were determined by counting of colony-forming units and compared to the specific requirements of the stand-alone test for contact lens disinfection of the hygiene standard ISO 14729. The concentration of both microorganisms was reduced by three orders of magnitude after less than 4 h of irradiation. The LED current and intensity have not yet been at maximum and could be further increased if necessary for other microorganisms. The presented device fulfils the requirement of the stand-alone test of the contact lens hygienic standard ISO 14729 for the tested Pseudomonas and Staphylococcus strains. According to literature data, the inactivation of Serratia marcescens , Candida albicans and Fusarium solani seems also possible, but may require increased LED current and intensity.

The aim of this study was to develop novel hydroxyapatite (HAP)-based bioactive bone replacement materials for segmental osteotomy reconstruction. Customized three-dimensional (3D) bone construct was manufactured from nanohydroxyapatite (nHAP) with poly(lactide-co-glycolide) (PLGA) coating using 3D models derived from the computed tomography (CT) scanning of the rabbit’s ulna and gradient 3D printing of the bone substitute mimicking the anatomical shape of the natural bone defect. Engineered construct revealed adequate micro-architectural design for successful bone regeneration having a total porosity of 64% and an average pore size of 256 μm. Radiography and micro-CT analysis depicted new bone apposition through the whole length of the reconstructed ulna with a small area of non-resorbed construct in the central area of defect. Histological analysis revealed new bone formation with both endochondral and endesmal type of ossification. Immunohistochemistry analysis depicted the presence of bone formation indicators – bone morphogenetic protein (BMP), osteocalcin (OCN) and osteopontin (OPN) within newly formed bone. Manufactured personalized construct acts as a “smart” responsive biomaterial capable of modulating the functionality and potential for the personalized bone reconstruction on a clinically relevant length scale.